1922] ARBER—LEAVES OF FARINOSAE . 87 
Philydraceae 
This small family consists of four species assigned to three 
genera. Sections were secured of three of these species, one 
belonging to each genus. The monotypic Philydrum lanuginosum 
Banks (fig. 22) has an ensiform leaf, whose shape and plan of 
vascular anatomy recall Anarthria scabra R.Br.; the leaf, however, 
is conspicuously lacunate. Pritzelia pygmaea F. Muell., repre- 
senting the second of these monotypic genera, is related to Phi- 
lydrum in its leaf structure, very much as Anarthria gracilis is 
related to A. scabra. The leaf of Pritzelia has a sheathing base 
(fig. 21 4) and a limb, which so closely approximates to radial 
structure that it is not possible to determine from internal evidence 
which is the median bundle (fig. 21 B). The leaf contains a 
number of large, solitary, acicular crystals (c, fig. 21C). The 
leaves of the third genus, Helmholtzia, have been described as 
equitant and ensiform, but Dr. Starr has been so kind as to inform 
me that in both the two species of which the genus consists, 
H. acorifolia F. Muell. and H. glaberrimum Hook. f., the vaginal 
portion of the leaf is strongly keeled, and that this keel runs as a 
midrib throughout the “vertical” limb. The limb is thus expanded 
in a plane at right angles to the truly ensiform limb of Philydrum. 
In accordance with this difference of construction, the leaf anatomy 
of the species which I have been able to examine, H. acorifolia, 
proves to be dorsiventral, with palisade parenchyma on the adaxial 
surface (fig. 23 A, B). The most striking feature of the leaf from 
the present standpoint is that, despite its dorsiventrality, it is 
characteristically phyllodic in structure, containing, besides 
normally orientated bundles (mb), others that are inverted (7b). 
Fig. 23 C shows the inverted group from the upper side of the 
midrib, in greater detail. 
Conclusions 
CLASSIFICATION OF LEAF TYPES IN FARINOSAE 
In the preceding pages the treatment of the leaves of this 
cohort has been almost exclusively descriptive, but I propose now 
to consider the interpretation of their morphology. The most 
